The proposed project develops and tests innovative methods to assess and increase the use of screening mammography for the early detection and treatment of breast cancer in older women in accordance with the recommendations of the National Cancer Institute and other national organizations. The project uses two methodologies: (a) the assessment of older women and of physicians regarding utilization of mammography through the analysis of Medicare claims data and (b) changing the behavior of older women who have not recently utilized mammography through the use of personal communications. These studies are performed in populations of older women in general, older African American women, and older women living in rural areas. The methods should be cost-effective: they use data already collected by Medicare; the intervention targets only individuals needing to change, communicating the need to change in inexpensive personal letters. The major study targets older women directly. In the first (assessment) component of this study, Medicare claims for older women are analyzed to identify women who have and who have not had a recent mammogram. The second (intervention) component is a randomized controlled trial focusing on older women who have not had a recent mammogram. Those assigned to an intervention group will be sent a personalized letter noting they have not had a recent mammogram and encouraging them to obtain one. The programmatic effectiveness and cost effectiveness of the intervention will be assessed. The third (associated factors) component examines the relationship of beneficiary characteristics to whether or not women originally had a mammogram and to whether or not women in the randomized controlled trial obtained a mammogram. All three components are performed on a representative population of older women and on representative subpopulations of older African American women and of older women living in rural areas. An additional study targets primary care physicians. In the first (assessment) component, Medicare data are used to identify the percent of a physician's Medicare patients who have had a recent mammogram. A second (associated factors) component examines relationships between physician characteristics (obtained from existing data sets) to a physician's score on percent of their patients screened. Both components are performed on a representative population of primary care physicians and on physicians treating higher than average proportions of older African American women and older women living in rural areas. When the first (assessment) component is successfully completed, randomized controlled interventions will be proposed that will use personalized letters sent to physicians with a low percent of screened older women patients. This project's methods can be applied to female Medicare beneficiaries and their physicians throughout the United States to assess and increase use of screening mammography. Additionally, the methods provide models for using Medicare claims to improve surveillance and dissemination regarding use of other services.